See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/221165001

Mapping strategies in DJ scratching.

Conference Paper · January 2006

Source: DBLP

CITATIONS READS

16 5,574

2 authors:

Kjetil Falkenberg Roberto Bresin

KTH Royal Institute of Technology KTH Royal Institute of Technology
48 PUBLICATIONS 294 CITATIONS 153 PUBLICATIONS 3,637 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Roberto Bresin on 19 May 2014.

The user has requested enhancement of the downloaded file.

 Proceedings of the 2006 International Conference on New Interfaces for Musical Expression (NIME06), Paris, France
Mapping strategies in DJ scratching
Kjetil Falkenberg Hansen
Dep of Speech, Music and Hearing, KTH
Stockholm, Sweden
kjetil@kth.se
ABSTRACT
For 30 years Disc Jockeys have been expressing their mu-
sical ideas with scratching. Unlike many other popular in-
struments, the equipment used for scratching is not built
as one single unit, and it was not intended to be a mu-
sical instrument. This paper gives an overview of how
DJs use their turntable, vinyl record and audio mixer in
junction to produce scratch music. Their gestural input
to the instrument is explained by looking at the mapping
principles between the controller parameters and the audio
output parameters. Implications are discussed for the de-
sign of new interfaces with examples of recent innovations
and experiments in the field.
Keywords
Instrument mapping, DJ, scratching, controllers, virtual
instruments
1. BACKGROUND
The success of a musical instrument will have a multi-
tude of reasons. Some immediate suggestions are that it
is cool to look at while somebody plays it; it sounds re-
ally good; one can express feelings and communicate emo-
tions with it; it is cheap; you can feel its vibrations; it has
blinking lights; it has a shoulder strap; you can hit it...
Scratching is one such successful “instrument” that takes
some effort to get acquainted with.
Scratching is, firstly, not an instrument, it is what Disc
Jockeys (DJs) sometimes do with turntables and mixers.
The popularity and history of this musical niche has been
explained in several aspects in previous works (see for in-
stance [18, 20, ?, 3, 9]). The combination of using both
a turntable and a mixer laid the ground for scratching
about three decades ago, but despite this, DJs only refer
to their instrument as “turntable”, inadvertently including
the mixer unit. Indeed, analysis of common DJ scratch-
ing techniques have shown that the mixer is almost as
important for the musician as the turntable [10]. Impera-
tive to the control and interface part is the sound source,
the vinyl record. Sound generation is therefore controlled
by two different devices, but dependent on what is cur-
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise,
or republish, to post on servers or to redistribute to lists, requires prior
specific permission and/or a fee.
NIME 06, June 4-8, 2006, Paris, France
Copyright remains with the author(s).
188
Roberto Bresin
Dep of Speech, Music and Hearing, KTH
Stockholm, Sweden
roberto@kth.se
rently playing on the record. In this sense, scratching can
be compared to new computer based instruments that has
separated the interface and controllers from the sound gen-
erator.
Recent alternative interfaces for scratching follow to a
great deal the separation of a sound level unit and a sound
’source’ unit. Why is this so? Should not the effort of
the new interface developers be focussed on the task of
making one instrument for scratching? Evidently, there
is a mismatch between development of the turntable-part
and the mixer/crossfader-part: this is because the struggle
to make an acceptable substitute to the turntable is of
highest priority. For instance, all leading manufacturers
have each marketed a number of devices for scratching
digitally stored sound, but only one of them has produced
a mixer with additional control possibilities [19].
2. CONTROLLERS AND MAPPINGS
Mappings in interfaces for scratching are quite intuitive,
but not simple one-to-one mappings as it may seem at a
first glance. One hand controls the playback speed of the
record by pushing and dragging it with fast, short move-
ments. The other hand controls the sound level out from
the system, in general only the extremes of total silence
and full sound1 .
2.1 Common interfaces
2.1.1 The turntable and vinyl record
Playback speed modulations directly affects the pitch
we hear, but they do more also. Sound samples2 used
in scratching are often words, short phrases, synthesized
sounds, instrumental sounds, and sound effects from movies
or daily life. However, they are seldom played as withheld
and straight tones of constant pitch. Therefore the per-
ception of the sound changes dramatically when played at
different speeds and backwards. This is true not only in
the time domain when sounds are compressed or stretched
out in time, but also in frequency domain when the whole
spectrum is shifted up or down. To disguise the origin of
the sound being scratched further, there is normally a fast
acceleration and deceleration with each hand gesture that
transforms the sound.
1
The DJ mixers also have tone controllers and linear am-
plitude sliders, but the most utilized control is the log-
arithmic crossfader which works as an on-off switch for
sound level.
2
In this text, samples are recorded sounds on vinyl records.
Samples can be short (from a tenth of a second) or long
(to more than a minute), but typically they are a bit less
than one second.
 Proceedings of the 2006 International Conference on New Interfaces for Musical Expression (NIME06), Paris, France
Another aspect to consider related to the mapping be-
tween the record-moving gesture and the sound we hear,
is clearly the sound source on the record itself. As stated,
pitch changes and artifacts in the time domain will oc-
cur in the interaction with the turntable. But these are
not at all consistent for all sounds. For instance, short
sounds of low frequency are hard to stretch out in order to
manipulate them with the crossfader, while longer sounds
easily allows virtuous use of crossfader. Some sounds are
unfit for playing with pitch, such as the sound of a snare
drum hit, while others such as a sample of James Brown
shouting ’aaayeah’, are perfect for that purpose.
Not only the sounds on the record are important for
identifying the mapping. Even the silent parts before and
after each sound sample must be taken into account. If
the DJ starts to move the record before the needle reaches
the spot where the sound starts, the sound will start with
a high pitch. If she starts the move when the needle is
located within the sound, a sudden glissando effect can
be heard. Likewise, if the DJ pulls the record back and
forth repeatedly over an entire short sound sample, we
will hear a succession of sharp tone onsets, while the same
movements “inside” for instance the long James Brown
sample will generate a very different, more siren-like sound
gliding fast up and down in pitch.
Consequently, if the record on the turntable consisted
of a long, unchanging sound, such as a synthesized tone
of constant pitch, the musician’s gesture in moving the
record would be more audible and apparent.
The three most common gestures for moving the record
are to place the hand flat down with fingertips on the
record and use the elbow to move; to hold the hand at an
angle and use the wrist; and to hold the hand steady and
only make small movements by using the finger joints. In
the first two cases, the hand and fingers are flexible and
can supplement the main gesture.
2.1.2 The audio mixer
The audio mixer is placed between the turntable and
the loudspeaker, so the signal from the record is always
going through the mixer. It has a number of controllers,
most important are the crossfader, a line/phono switch, a
volume fader and one or more knobs for equalization (tone
control). DJs use the crossfader much more than the other
mixer controllers. During recent years, the crossfader has
evolved from fading linearly between the two turntables in
the DJ set-up to being a logarithmic fader with very steep
fading curves that in practice is only an on-off switch for
sound. Traditionally, the line/phono switch was used for
turning the sound quickly and sharply on and off, but
today it is superseded by the crossfader, which is much
better suited to fast manipulations than the switch.
The crossfader knob can be moved using different hand
gestures, but two methods are dominant. The knob is
either pinched between the thumb and a finger and pushed
the short distance on the fader that runs from silent to
sound, normally adjusted to 1-2 millimeters, or it can be
lightly hit with one or more fingers against the thumb
which acts as a spring, bouncing the fader back. Either
way, the starting position can be both in the silent and the
full sound area. The crossfader can be reversed so that the
silent part can be on either the extreme right or the left
side of the fader run.
2.2 Current mapping in scratching
By following principles for describing instrument map-
189
ping strategies suggested by Hunt et al. (in [12, 13, 14])
an overview of the existing mapping between gestures and
sound in scratching is presented here. The controller pa-
rameters considered are record speed, sound source (on
the record), the needle’s placement in the sound groove,
and on the mixer the use of crossfader, volume slider and
tone control knobs. The audio output parameters consid-
ered are pitch (not necessarily in the meaning of discrete
semitone steps), tone onsets and durations, and tone or
sound characteristic such as timbre and dynamics. Fig-
ure 1 shows the relations between controller and audio
output parameters.
2.2.1 Timbre and dynamics
Timbre is determined by the sound source used, and the
speed at which it is played back. In addition, the DJ can
control timbre to a certain degree using the tone control
knobs, which is done for certain scratch techniques [9].
Dynamics, or sound level variation, is a very important
factor for expressive playing in most musical instruments.
However, it is rarely found in scratching. Sound level is
determined by the volume slider on the mixer, the sound
source and the playback speed. The crossfader can not be
regarded as a controller for dynamics, as its fading curve is
far too steep. The needle’s placement in the sound groove
can have a great effect on dynamics, as many sounds de-
crease in amplitude over time.
Timbral and dynamical variations can also be achieved
by choosing locations with different amount of wearing
on the record. In records where one sound is repeated
(e.g. for several minutes, as in [7]) the sound can be more
detoriated in some grooves than others, making the same
sample sound different depending on where the needle is
located [11].
2.2.2 Pitch
Pitch is determined by the sound source and is adjusted
straightforwardly with the playback speed. Experienced
DJs sometimes use the pitch adjustment slider on the
turntable to “tune” the sound sample to the tonality of
the piece to which they are scratching. A large part of all
(scratched) tones have the same speed as, or close to, the
revolving turntable [10]. For many of the favored sounds
on the record, it is hard to hear a fundamental frequency.
Several samples, for instance the popular “Fresh” (a long
shhh sound, [8]), have a noise band with a broad maxi-
mum, inducing the perception of some pitch.
2.2.3 Tone onsets and durations
By tone onsets it is meant both the start of a tone and
the characteristics of its attack. Attack properties are
controlled by record speed, sound placement and sound
source. A slow push produces a slow rise in pitch and a
“soft” attack, while a fast push produces a harder attack.
A movement starting with the needle positioned inside a
sample produces a softer attack than the same movement
starting before the sample starts. The sample of a snare
drum hit as sound source will have a very different attack
from a sample of a moaning soul singer.
The rapid stream of tone attacks is a very noticeable
and distinguishing aspect of scratching. In a typical tech-
nique with a duration a couple of tenths of a second, there
can be as many as 6-9 tone attacks [10]. The most im-
portant controller for generating such rapid onsets is the
crossfader, especially when the fingers bounce it against
the thumb. Also quick shifts in playback tempo from go-
 Proceedings of the 2006 International Conference on New Interfaces for Musical Expression (NIME06), Paris, France

record
note characteristics sound sample

duration note characteristics

record movement
pitch
turntable
needle placement
timbre
onsets

dynamics crossfader
/line switch mixer

sound characteristics tone controllers sound characteristics

volume slider

Figure 1: Mapping sound output parameters (left and right) to control parameters (middle) in a normal
instrument setup for scratching. Onset and pitch are undeniably the most crucial sound attributes, and
are placed to the right in the figure. Line thickness (from dashed to bold lines) represent the importance
of the connection. The bold boxes are considered to be the most essential control parameters: record
movement, sound sample properties and crossfader movement.

ing forward to going backward generates onsets. Certain
techniques can produce more than 20 onsets a second only
using the record speed. A third control possibility is to
use sound sources where short sounds are placed close to
each other, for instance a slow drum roll or hand-clapping.
One single gesture of moving the record over such samples
produces a multitude of tone attacks.
The length of the sound sample influences the durations
of the tones which can be controlled by hand gestures on
the turntable. The crossfader can at any time shorten the
duration by stopping the tone. It is often hard to sustain
short samples for long, which is also still the weakest point
of all the interfaces for scratching digital sound files. Even
long samples can be hard to sustain with a constant speed,
as the turntable’s tone arm obstructs and makes circular
gestures problematic.
2.3 New interfaces for scratching
New interfaces introduce new possibilities. Generally,
the common first feature is to give control over an addi-
tional parameter, either in the sound itself or in processing
the sound [16]. With scratching, the first step has been
to reproduce the existing hardware (the turntable part)
in software models and with alternative controllers, often
realized as some type of a revolving wheel resembling the
turntable. This strategy is obviously in conflict with the
design principles lined up by Cook [6]. The pursuit to pro-
duce an acceptable virtual turntable has led to commer-
cially available interfaces for controlling sound files that
are quite good. Solutions for allowing control over ad-
ditional parameters, however, is still a rather unexplored
field, and the big companies have not yet acknowledged
the potential.
In previous NIMEs and in related journals, and even in
commercially available products, several new interfaces for
DJs have been envisaged. This section of the paper gives
a short review of possible mapping strategies of a selection
of these interfaces with regards to the previous section.
Of the many interfaces manipulating digitally stored
sounds, Final Scratch by Stanton has probably been most
accepted, using a time-coded vinyl record to control sound
files on a computer. Other such interfaces include devices
for scratching with CDs, marketed in various designs by
all leading manufacturers (e.g. Pioneer, Technics, Stanton,
Vestax and Denon). Common for this group of interfaces
is that they do not offer new ideas for controllers nor new
mapping possibilities.
DVD scratching is a similar concept to above mentioned
scratch tools for digital media. Although few companies
have marketed instruments for this, Pioneer being one ex-
ception [17], it is a promising performance concept for VJs
(video DJs). Current version can treat the video image
equivalently to sound by modulating the playback speed.
The visuals introduce new output parameters for the inter-
face, which parameters can be mapped to other controllers
than just the ’turntable’ gestures.
Mixxx by Andersen [1] is developed both as a set of
performance tools and as a means of studying DJ perfor-
mances. Both mixer and turntable interfaces are replicas
of the standard interfaces, but they also send MIDI mes-
sages. For performance, it enables to have a mapping layer
between the controller parameters and the audio output,
so that they can be used for controlling new or existing
sound properties. Mixxx also includes a new visual feed-
back technique in an augmented turntable, the Fisheye [2],
a method for projecting a waveform image on the record.
In the presented implementation, Mixxx is not addressing
DJ scratching specifically.
Mixxx also includes a new visual feedback technique in
an augmented turntable, the Fisheye[2], a method for pro-
jecting a sound image on the record.
The D’Groove by Beamish et al. (in [5] and [4] intro-
duces haptics, induced by a force feedback motor, as a
feedback method in the turntable interface. This allows for
haptic ’visualization’ of the sounds on the record, which
assists the DJ in for instance finding the sound sample
’borders’. In tests, experienced scratch DJs quickly start
to experiment with new techniques and performance ideas,
and learned to use the novel features that haptic feedback
provided. Some new controllers were also implemented,
but did not suggest new mapping layers or strategies, as
was the case with Mixxx.
Skipproof [11] is a software tool for evaluating synthe-
sized scratch techniques, and also a scratch performance

190
 Proceedings of the 2006 International Conference on New Interfaces for Musical Expression (NIME06), Paris, France
instrument. The controller and audio output parameters
found in common DJ set-ups can be mapped freely to var-
ious devices for computer input, such as MIDI interfaces
or gesture sensors. Skipproof does not have the general
approach to DJing as the above mentioned interfaces, but
is focussing on the skip proof feature found in special-
ized records for DJs where a one revolution long sample
(1.8 seconds) is looped. This allows for designing specific
scratch interfaces that contrast the established ones.
16padjoystickcontroller by Lippit [15] is a realtime sam-
pling system for scratch DJs. Lippit acknowledges that
DJs are fully occupied with their hands during perfor-
mance and suggest supplementary controllers, a foot switch,
a joystick and a drum pad, that enable the DJ to in-
teract with their own performances in real-time. With
these controllers, DJs can record and manipulate portions
of their scratch performance. In such context, the added
controllers are separated from the existing ones and repre-
sent isolated instrument systems without possibilities for
interaction.
Samurai mixers by Vestax [19] are the only commer-
cially available tools from a major manufacturer that have
added possibilities for sound control with the crossfader.
The crossfader is programmable so that the DJ can han-
dle amplitude patterns instead of the normal fading curve,
so that one gesture on the crossfader can generate several
onsets, not just one. Vestax has also marketed a turntable
with integrated crossfader, the QFO, but this interface
does not introduce new control possibilities or mappings.
It illustrates however the need for a dedicated scratch in-
strument.
3. CONCLUSIONS
The existing mappings between gestures and sound with
a turntable and a mixer should be considered when de-
signing and building new interfaces for scratching. Up to
now, gestures used for scratching have originated from the
idea of reading a sound file at varying speeds and direc-
tion, with a one-to-one mapping between the extent of a
(hand) movement and the position in a sound. Regardless
of the kind of gesture, it is always a measure of distance.
Gestures for controlling crossfader, or sound on/off, are
not as intuitive as the turntable-controlling gesture, and
also much less studied.
In general, commercial manufacturers of equipment and
instruments focus on existing control paradigms instead of
exploring new possibilities at hand. Immediate responses
from DJs participating in the described interface develop-
ments and experiments, indicate an openness to new con-
trol concepts and mappings for scratch performances. New
interfaces will probably contribute to develop the playing
styles and DJ techniques in other directions.
DJs have with great consensus developed a common lan-
guage of playing techniques that require complex, synchro-
nized gestures to be performed correctly [9]. Despite the
rather unusual and hard-to-master handling of the instru-
ment, its principles can be applied to other interfaces, both
for music performance and for general human-computer
interaction.
4. ACKNOWLEDGEMENTS
This work is supported by S2S2 (acronym for Sound
to Sense, Sense to Sound, pronounce s two - s square),
a FET-Open Coordination Action (number of contract:
IST-2004-03773).
191
View publication stats
5. REFERENCES
[1] T. H. Andersen. Mixxx: Towards Novel DJ
Interfaces. In Proceedings of the 2003 Conference on
New Interfaces for Musical Expression, pages 30–35,
Montreal, Canada, 2003.
[2] T. H. Andersen and K. Erleben. Sound interaction
by use of comparative visual displays. In Proceedings
of the Second Danish HCI Symposium. HC Tryk,
November 2002.
[3] T. Beamish. A Taxonomy of DJs. 2001.
[4] T. Beamish, K. MacLean, and S. Fels. Manipulating
Music: Multimodal Interaction for DJs. In
Proceedings of the Conference for Human-Computer
Interaction 2004, Vienna, Austria, April 2004.
[5] T. Beamish, K. van den Doel, K. MacLean, and
S. Fels. DGroove: A Haptic Turntable For Digital
Audio Control. In Proceedings of the 2003
International Conference on Auditory Display,
Boston, 2003.
[6] P. Cook. Principles for Designing Computer Music
Controllers. In Proceedings of the CHI 00, NIME
Workshop, 2000.
[7] DJ 1210 Jazz. Book of Five Scratches. Book 2.
Snickars Rec., SR1206, 2001.
[8] Fab Five Freddie. Change The Beat. 12” Celluloid
Records CEL 156, 1982.
[9] K. F. Hansen. The Basics of Scrathing. Journal of
New Music Research, 31(4):357–365, 2002.
[10] K. F. Hansen and R. Bresin. Analysis of a Genuine
Scratch Performance. In Proceedings of the Gesture
Workshop, pages 519–528, 2003.
[11] K. F. Hansen and R. Bresin. The Sounding Object,
chapter Complex Gestural Audio Control: The Case
of Scratching, pages 221–269. Mondo Estremo,
Florence, Italy, 2003.
[12] A. Hunt and R. Kirk. Multiparametric control of
real-time systems. In J. R. . M. W. Marc Battier,
editor, Trends in Gestural Control of Music. FSU -
Florida, April 2000.
[13] A. Hunt and R. Kirk. Trends in Gestural Control of
Music, chapter Mapping Strategies for Musical
Performance, pages 231–258. Ircam - Centre
Pompidou, 2000.
[14] A. Hunt, M. M. Wanderley, and M. Paradis. The
importance of parameter mapping in electronic
instrument design. Journal of New Music Research,
32(4):429–440, 2003.
[15] R. Khazam. Electroacoustic alchemist. The Wire
Magazine, 160:36–, 1997.
[16] T. M. Lippit. Realtime Sampling System for the
Turntablist, Version 2: 16padjoystickcontroller. In
Proceedings of the 2004 Conference on New
Interfaces for Musical Expression, 2004.
[17] J. A. Paradiso and S. O’Modhrain. Current Trends
in Electronic Music Interfaces. Journal of New
Music Research, 32(4):345–349, 2003.
[18] Pioneer. Pioneer DVJ-X1, 2003.
[19] D. Toop. Rap Attack. South End Press, Boston,
USA, 1984.
[20] Vestax. Vestax Samurai Mixer Series, 2006.
[21] M. White. The phonographic turntable and
performance practice in hip hop music. Technical
report, 1999.